{
  "openapi": "3.0.1",
  "info": {
    "title": "Website Contact Scraper: Email, Phone, Social",
    "description": "One row per website, not per page: emails, phone numbers in E.164, 31 social platforms, postal address, opening hours and VAT id. The mail domain is MX-checked, Cloudflare-obfuscated addresses are decoded, and no browser or proxy is involved.",
    "version": "0.1",
    "x-build-id": "fkvvcJNdgqMiLpD7I"
  },
  "servers": [
    {
      "url": "https://api.apify.com/v2"
    }
  ],
  "paths": {
    "/acts/snow_leo_data~website-contact-email-phone-social-scraper-lead-extractor/run-sync-get-dataset-items": {
      "post": {
        "operationId": "run-sync-get-dataset-items-snow_leo_data-website-contact-email-phone-social-scraper-lead-extractor",
        "x-openai-isConsequential": false,
        "summary": "Executes an Actor, waits for its completion, and returns Actor's dataset items in response.",
        "tags": [
          "Run Actor"
        ],
        "requestBody": {
          "required": true,
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/inputSchema"
              }
            }
          }
        },
        "parameters": [
          {
            "name": "token",
            "in": "query",
            "required": true,
            "schema": {
              "type": "string"
            },
            "description": "Enter your Apify token here"
          }
        ],
        "responses": {
          "200": {
            "description": "OK"
          }
        }
      }
    },
    "/acts/snow_leo_data~website-contact-email-phone-social-scraper-lead-extractor/runs": {
      "post": {
        "operationId": "runs-sync-snow_leo_data-website-contact-email-phone-social-scraper-lead-extractor",
        "x-openai-isConsequential": false,
        "summary": "Executes an Actor and returns information about the initiated run in response.",
        "tags": [
          "Run Actor"
        ],
        "requestBody": {
          "required": true,
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/inputSchema"
              }
            }
          }
        },
        "parameters": [
          {
            "name": "token",
            "in": "query",
            "required": true,
            "schema": {
              "type": "string"
            },
            "description": "Enter your Apify token here"
          }
        ],
        "responses": {
          "200": {
            "description": "OK",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/runsResponseSchema"
                }
              }
            }
          }
        }
      }
    },
    "/acts/snow_leo_data~website-contact-email-phone-social-scraper-lead-extractor/run-sync": {
      "post": {
        "operationId": "run-sync-snow_leo_data-website-contact-email-phone-social-scraper-lead-extractor",
        "x-openai-isConsequential": false,
        "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.",
        "tags": [
          "Run Actor"
        ],
        "requestBody": {
          "required": true,
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/inputSchema"
              }
            }
          }
        },
        "parameters": [
          {
            "name": "token",
            "in": "query",
            "required": true,
            "schema": {
              "type": "string"
            },
            "description": "Enter your Apify token here"
          }
        ],
        "responses": {
          "200": {
            "description": "OK"
          }
        }
      }
    }
  },
  "components": {
    "schemas": {
      "inputSchema": {
        "type": "object",
        "properties": {
          "startUrls": {
            "title": "Websites",
            "type": "array",
            "description": "Websites to read. A bare domain (<code>example.com</code>), a full link, or a link-in-bio page such as <code>linktr.ee/name</code> all work. One row comes back per address, even when the site refuses the request - the row then says why.",
            "default": [
              {
                "url": "https://www.austinfilm.org"
              },
              {
                "url": "https://jeffreysofaustin.com"
              },
              {
                "url": "https://www.yarddog.com"
              }
            ],
            "items": {
              "type": "object",
              "required": [
                "url"
              ],
              "properties": {
                "url": {
                  "type": "string",
                  "title": "URL of a web page",
                  "format": "uri"
                }
              }
            }
          },
          "maxItems": {
            "title": "Maximum rows",
            "minimum": 0,
            "maximum": 100000,
            "type": "integer",
            "description": "Hard stop on how many rows are written and charged. <code>0</code> means no limit. Sites beyond the limit are never opened, so the limit saves time as well as money.",
            "default": 100
          },
          "maxPagesPerSite": {
            "title": "Maximum pages per site",
            "minimum": 1,
            "maximum": 40,
            "type": "integer",
            "description": "How many pages of one site may be opened. Measured on 135 live business sites: reading only the home page finds an email on far fewer sites than reading the contact and imprint pages as well. Six is a good balance; raise it for large sites.",
            "default": 6
          },
          "crawlStrategy": {
            "title": "Which pages to open",
            "enum": [
              "contact-pages",
              "all-links",
              "homepage-only"
            ],
            "type": "string",
            "description": "<b>contact-pages</b> ranks links by how likely they lead to contacts (contact, imprint, about, team, locations) and opens the best ones. <b>all-links</b> opens same-site links in page order. <b>homepage-only</b> reads one page.",
            "default": "contact-pages"
          },
          "useSitemap": {
            "title": "Use sitemap.xml",
            "type": "boolean",
            "description": "When the home page has no visible link to a contact page - common on sites whose menu is drawn by a script - look for one in <code>robots.txt</code> and <code>sitemap.xml</code>.",
            "default": true
          },
          "guessContactPaths": {
            "title": "Try the usual contact paths",
            "type": "boolean",
            "description": "If nothing contact-like was found in the links or the sitemap, try well-known paths such as <code>/contact</code>, <code>/kontakt</code> and <code>/impressum</code>.",
            "default": true
          },
          "sameDomain": {
            "title": "Stay within the site",
            "type": "boolean",
            "description": "Follow only links that stay on the same domain. Link-in-bio pages are the deliberate exception: their whole point is to lead elsewhere.",
            "default": true
          },
          "followLinkInBioPages": {
            "title": "Follow link-in-bio pages",
            "type": "boolean",
            "description": "For <code>linktr.ee</code>, <code>bio.link</code>, <code>beacons.ai</code> and ten similar hosts, read the linked sites instead of the bio page itself, which has no contacts of its own.",
            "default": true
          },
          "scrapeIframes": {
            "title": "Read iframes",
            "type": "boolean",
            "description": "Also open frames embedded in the page. Some site builders put the footer with the address into a frame. Off by default because frames often hold adverts.",
            "default": false
          },
          "checkMx": {
            "title": "Check the mail domain (MX)",
            "type": "boolean",
            "description": "Ask DNS whether each email domain can receive mail at all, and name the mail provider. A missing MX and A record proves the address is undeliverable; a present MX does not prove the mailbox exists, and this Actor never claims it does.",
            "default": true
          },
          "defaultCountry": {
            "title": "Fallback country for phone numbers",
            "type": "string",
            "description": "Two-letter country code used to build the international <code>+</code> form when the site itself gives no clue - no country domain and no address in its structured data. Leave empty to keep such numbers raw rather than guess."
          },
          "onlyWithEmail": {
            "title": "Only sites with an email",
            "type": "boolean",
            "description": "Drop rows where no email was found.",
            "default": false
          },
          "onlyWithPhone": {
            "title": "Only sites with a phone number",
            "type": "boolean",
            "description": "Drop rows where no phone number was found, in any form.",
            "default": false
          },
          "onlyWithContacts": {
            "title": "Only sites with any contact",
            "type": "boolean",
            "description": "Keep a row if it has an email, a phone number or a social profile.",
            "default": false
          },
          "onlyWithSocialProfile": {
            "title": "Only sites with a social profile",
            "type": "boolean",
            "description": "Drop rows where no social media profile was found on the site.",
            "default": false
          },
          "skipUnreachableSites": {
            "title": "Skip sites that did not answer",
            "type": "boolean",
            "description": "By default a site that refuses the request or no longer exists still gets a row, so your list stays the same length and says what happened to each address. Switch this on to drop those rows instead - and not pay for them.",
            "default": false
          },
          "onlyDeliverableEmail": {
            "title": "Only sites with a deliverable email domain",
            "type": "boolean",
            "description": "Keep a row only if at least one email sits on a domain that has an MX (or A) record. Needs <b>Check the mail domain</b> switched on - with it off the run stops before opening anything and says so, instead of quietly returning nothing.",
            "default": false
          },
          "onlySameDomainEmail": {
            "title": "Only emails on the site's own domain",
            "type": "boolean",
            "description": "Drop addresses hosted elsewhere - an agency mailbox, a marketplace, a web designer's own address in the footer.",
            "default": false
          },
          "skipFreeMailboxes": {
            "title": "Skip free mailboxes",
            "type": "boolean",
            "description": "Drop gmail.com, gmx.de, outlook.com and 50 other consumer mail hosts.",
            "default": false
          },
          "emailKinds": {
            "title": "Keep only these kinds of address",
            "type": "array",
            "description": "<b>role</b> - info@, sales@, support@ and 70 more. <b>personal</b> - looks like a person's name. <b>generic</b> - anything else. <b>noreply</b> - automated senders. Empty means keep all kinds.",
            "items": {
              "type": "string",
              "enum": [
                "role",
                "personal",
                "generic",
                "noreply"
              ],
              "enumTitles": [
                "Role mailboxes",
                "Personal",
                "Generic",
                "No-reply"
              ]
            },
            "default": []
          },
          "domainContains": {
            "title": "Domain contains",
            "type": "array",
            "description": "Keep only sites whose domain contains one of these words.",
            "default": [],
            "items": {
              "type": "string"
            }
          },
          "textContains": {
            "title": "Page text contains",
            "type": "array",
            "description": "Keep only sites whose title, description or company name contains one of these words.",
            "default": [],
            "items": {
              "type": "string"
            }
          },
          "excludeContains": {
            "title": "Exclude when it contains",
            "type": "array",
            "description": "Drop sites whose address, title, description or company name contains one of these words.",
            "default": [],
            "items": {
              "type": "string"
            }
          },
          "onlyChanged": {
            "title": "Monitor mode: only what changed",
            "type": "boolean",
            "description": "Remember what each site returned and, on the next run, write only the sites whose contacts changed. Nothing changed means an empty dataset and no charge - the log says so in plain words. Memory lives in a named key-value store, so it survives between runs.",
            "default": false
          },
          "emitUnchanged": {
            "title": "Emit unchanged sites too",
            "type": "boolean",
            "description": "In monitor mode, write unchanged sites as well, tagged <code>UNCHANGED</code>.",
            "default": false
          },
          "monitorStoreName": {
            "title": "Memory store name",
            "type": "string",
            "description": "Name of the key-value store holding the monitor memory. Use different names to keep several lists apart.",
            "default": "website-contacts-state"
          },
          "compactOutput": {
            "title": "Compact output",
            "type": "boolean",
            "description": "Drop empty fields and the per-address detail arrays. Smaller rows for AI agents and spreadsheets.",
            "default": false
          },
          "concurrency": {
            "title": "Sites in parallel",
            "minimum": 1,
            "maximum": 12,
            "type": "integer",
            "description": "How many sites are read at the same time.",
            "default": 6
          },
          "requestTimeoutSecs": {
            "title": "Request timeout, seconds",
            "minimum": 5,
            "maximum": 120,
            "type": "integer",
            "description": "How long to wait for one page before giving up on it. Small business sites on cheap hosting are slow; twenty seconds covered every site that answered at all in the 135-site measurement.",
            "default": 20
          }
        }
      },
      "runsResponseSchema": {
        "type": "object",
        "properties": {
          "data": {
            "type": "object",
            "properties": {
              "id": {
                "type": "string"
              },
              "actId": {
                "type": "string"
              },
              "userId": {
                "type": "string"
              },
              "startedAt": {
                "type": "string",
                "format": "date-time",
                "example": "2025-01-08T00:00:00.000Z"
              },
              "finishedAt": {
                "type": "string",
                "format": "date-time",
                "example": "2025-01-08T00:00:00.000Z"
              },
              "status": {
                "type": "string",
                "example": "READY"
              },
              "meta": {
                "type": "object",
                "properties": {
                  "origin": {
                    "type": "string",
                    "example": "API"
                  },
                  "userAgent": {
                    "type": "string"
                  }
                }
              },
              "stats": {
                "type": "object",
                "properties": {
                  "inputBodyLen": {
                    "type": "integer",
                    "example": 2000
                  },
                  "rebootCount": {
                    "type": "integer",
                    "example": 0
                  },
                  "restartCount": {
                    "type": "integer",
                    "example": 0
                  },
                  "resurrectCount": {
                    "type": "integer",
                    "example": 0
                  },
                  "computeUnits": {
                    "type": "integer",
                    "example": 0
                  }
                }
              },
              "options": {
                "type": "object",
                "properties": {
                  "build": {
                    "type": "string",
                    "example": "latest"
                  },
                  "timeoutSecs": {
                    "type": "integer",
                    "example": 300
                  },
                  "memoryMbytes": {
                    "type": "integer",
                    "example": 1024
                  },
                  "diskMbytes": {
                    "type": "integer",
                    "example": 2048
                  }
                }
              },
              "buildId": {
                "type": "string"
              },
              "defaultKeyValueStoreId": {
                "type": "string"
              },
              "defaultDatasetId": {
                "type": "string"
              },
              "defaultRequestQueueId": {
                "type": "string"
              },
              "buildNumber": {
                "type": "string",
                "example": "1.0.0"
              },
              "containerUrl": {
                "type": "string"
              },
              "usage": {
                "type": "object",
                "properties": {
                  "ACTOR_COMPUTE_UNITS": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATASET_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATASET_WRITES": {
                    "type": "integer",
                    "example": 0
                  },
                  "KEY_VALUE_STORE_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "KEY_VALUE_STORE_WRITES": {
                    "type": "integer",
                    "example": 1
                  },
                  "KEY_VALUE_STORE_LISTS": {
                    "type": "integer",
                    "example": 0
                  },
                  "REQUEST_QUEUE_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "REQUEST_QUEUE_WRITES": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATA_TRANSFER_INTERNAL_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATA_TRANSFER_EXTERNAL_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "PROXY_RESIDENTIAL_TRANSFER_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "PROXY_SERPS": {
                    "type": "integer",
                    "example": 0
                  }
                }
              },
              "usageTotalUsd": {
                "type": "number",
                "example": 0.00005
              },
              "usageUsd": {
                "type": "object",
                "properties": {
                  "ACTOR_COMPUTE_UNITS": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATASET_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATASET_WRITES": {
                    "type": "integer",
                    "example": 0
                  },
                  "KEY_VALUE_STORE_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "KEY_VALUE_STORE_WRITES": {
                    "type": "number",
                    "example": 0.00005
                  },
                  "KEY_VALUE_STORE_LISTS": {
                    "type": "integer",
                    "example": 0
                  },
                  "REQUEST_QUEUE_READS": {
                    "type": "integer",
                    "example": 0
                  },
                  "REQUEST_QUEUE_WRITES": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATA_TRANSFER_INTERNAL_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "DATA_TRANSFER_EXTERNAL_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "PROXY_RESIDENTIAL_TRANSFER_GBYTES": {
                    "type": "integer",
                    "example": 0
                  },
                  "PROXY_SERPS": {
                    "type": "integer",
                    "example": 0
                  }
                }
              }
            }
          }
        }
      }
    }
  }
}